mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Andrew Morton <akpm@linux-foundation.org>, Andi Kleen <ak@suse.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
	Chris Wright <chrisw@sous-sol.org>,
	virtualization@lists.osdl.org,
	lkml <linux-kernel@vger.kernel.org>,
	Xen-devel <xen-devel@lists.xensource.com>
Subject: [patch 24/33] xen: xen: hack to prevent bad segment register reload
Date: Tue, 22 May 2007 15:10:05 +0100	[thread overview]
Message-ID: <20070522141253.058860996@goop.org> (raw)
In-Reply-To: <20070522140941.802382212@goop.org>

[-- Attachment #1: xen-segreg-hack.patch --]
[-- Type: text/plain, Size: 1593 bytes --]

The hypervisor saves and restores the segment registers as part of the
state is saves while context switching.  If, during a context switch,
the next process doesn't use the TLS segments, it invalidates the GDT
entry, causing the segment register reload to fault.  This fault
effectively doubles the cost of a context switch.

This patch is a band-aid workaround which clears the usermode %gs
after it has been saved for the previous process, but before it gets
reloaded for the next, and it avoids having the hypervisor attempt to
erroneously reload it.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>

---
 arch/i386/xen/enlighten.c |   12 ++++++++++++
 1 file changed, 12 insertions(+)

===================================================================
--- a/arch/i386/xen/enlighten.c
+++ b/arch/i386/xen/enlighten.c
@@ -291,6 +291,18 @@ static void xen_load_tls(struct thread_s
 	load_TLS_descriptor(t, cpu, 2);
 
 	xen_mc_issue(PARAVIRT_LAZY_CPU);
+
+	/*
+	 * XXX sleazy hack: If we're being called in a lazy-cpu zone,
+	 * it means we're in a context switch, and %gs has just been
+	 * saved.  This means we can zero it out to prevent faults on
+	 * exit from the hypervisor if the next process has no %gs.
+	 * Either way, it has been saved, and the new value will get
+	 * loaded properly.  This will go away as soon as Xen has been
+	 * modified to not save/restore %gs for normal hypercalls.
+	 */
+	if (xen_get_lazy_mode() == PARAVIRT_LAZY_CPU)
+		loadsegment(gs, 0);
 }
 
 static void xen_write_ldt_entry(struct desc_struct *dt, int entrynum, u32 low, u32 high)

-- 


  parent reply	other threads:[~2007-05-22 15:23 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-05-22 14:09 [patch 00/33] xen: Xen paravirt_ops implementation Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 01/33] xen: paravirt: add an "mm" argument to alloc_pt Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 02/33] xen: paravirt: add a hook for once the allocator is ready Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 03/33] xen: paravirt: increase IRQ limit Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 04/33] xen: paravirt: unstatic leave_mm Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 05/33] xen: paravirt: unstatic smp_store_cpu_info Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 06/33] xen: paravirt: make siblingmap functions visible Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 07/33] xen: paravirt: export __supported_pte_mask Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 08/33] xen: Allocate and free vmalloc areas Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 09/33] xen: Add nosegneg capability to the vsyscall page notes Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 10/33] xen: Add Xen interface header files Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 11/33] xen: Core Xen implementation Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 12/33] xen: Xen virtual mmu Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 13/33] xen: xen event channels Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 14/33] xen: xen time implementation Jeremy Fitzhardinge
2007-06-06  8:39   ` [Xen-devel] " Jan Beulich
2007-06-06  8:54     ` Keir Fraser
2007-06-06  9:30       ` Jan Beulich
2007-06-06  9:56         ` Keir Fraser
2007-06-06 11:00           ` Jan Beulich
2007-06-06 11:52             ` Keir Fraser
2007-06-06 12:18               ` Andi Kleen
2007-06-06 12:46                 ` Jan Beulich
2007-06-06 12:53                   ` Andi Kleen
2007-06-06 12:54                   ` Keir Fraser
2007-06-06 10:05     ` Jeremy Fitzhardinge
2007-06-06 10:20       ` Jan Beulich
2007-06-06 10:26       ` Andi Kleen
2007-06-06 14:15         ` Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 15/33] xen: xen configuration Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 16/33] xen: xen: add pinned page flag Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 17/33] xen: Complete pagetable pinning for Xen Jeremy Fitzhardinge
2007-05-22 14:09 ` [patch 18/33] xen: xen: ignore RW mapping of RO pages in pagetable_init Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 19/33] xen: Account for time stolen by Xen Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 20/33] xen: Implement xen_sched_clock Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 21/33] xen: Xen SMP guest support Jeremy Fitzhardinge
2007-06-06  7:34   ` [Xen-devel] " Jan Beulich
2007-06-06  8:33     ` Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 22/33] xen: Add support for preemption Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 23/33] xen: xen: lazy-mmu operations Jeremy Fitzhardinge
2007-05-22 14:10 ` Jeremy Fitzhardinge [this message]
2007-05-22 14:10 ` [patch 25/33] xen: Use the hvc console infrastructure for Xen console Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 26/33] xen: Add Xen grant table support Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 27/33] xen: Add the Xenbus sysfs and virtual device hotplug driver Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 28/33] xen: Add Xen virtual block device driver Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 29/33] xen: Add the Xen virtual network " Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 30/33] xen: Xen machine operations Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 31/33] xen: xen: handle external requests for shutdown, reboot and sysrq Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 32/33] xen: Place vcpu_info structure into per-cpu memory, if possible Jeremy Fitzhardinge
2007-05-22 14:10 ` [patch 33/33] xen: Attempt to patch inline versions of common operations Jeremy Fitzhardinge
2007-05-22 14:57 ` [patch 00/33] xen: Xen paravirt_ops implementation Andi Kleen
2007-05-22 15:06   ` Jeremy Fitzhardinge

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20070522141253.058860996@goop.org \
    --to=jeremy@goop.org \
    --cc=ak@suse.de \
    --cc=akpm@linux-foundation.org \
    --cc=chrisw@sous-sol.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=torvalds@linux-foundation.org \
    --cc=virtualization@lists.osdl.org \
    --cc=xen-devel@lists.xensource.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®